Download Mol Microbiol. Jun 21. [Epub ahead of print]

Molecular Microbiology (2010) 77(4), 930–942 !
doi:10.1111/j.1365-2958.2010.07259.x
First published online 8 July 2010
H-NOX regulation of c-di-GMP metabolism and biofilm
formation in Legionella pneumophila
mmi_7259 930..942
Hans K. Carlson,1 Russell E. Vance1,2 and
Michael A. Marletta1,2*
Departments of 1Chemistry and 2Molecular and Cell
Biology, University of California, Berkeley, CA 94720,
USA.
Summary
Haem Nitric oxide/OXygen (H-NOX) binding domains
are a family of haemoprotein sensors that are widespread in bacterial genomes, but limited information
is available on their function. Legionella pneumophila is the only prokaryote found, thus far, to
encode two H-NOX proteins. This paper presents
data supporting a role for one of the L. pneumophila
H-NOXs in the regulation of biofilm formation. In
summary: (i) unmarked deletions in the hnox1 gene
do not affect growth rate in liquid culture or replication in permissive macrophages; (ii) the Dhnox1
strain displays a hyper-biofilm phenotype; (iii) the
gene adjacent to hnox1 is a GGDEF-EAL protein,
lpg1057, and overexpression in L. pneumophila of
this protein, or the well-studied diguanylate cyclase,
vca0956, results in a hyper-biofilm phenotype; (iv)
the Lpg1057 protein displays diguanylate cyclase
activity in vitro and this activity is inhibited by the
Hnox1 protein in the Fe(II)-NO ligation state, but not
the Fe(II) unligated state; and (v) consistent with the
Hnox1 regulation of Lpg1057, unmarked deletions of
lpg1057 in the Dhnox1 background results in reversion of the hyper-biofilm phenotype back to wildtype biofilm levels. Taken together, these results
suggest a role for hnox1 in regulating c-di-GMP production by lpg1057 and biofilm formation in
response to NO.
Introduction
The Haem-Nitric oxide/OXygen (H-NOX) binding domain
is best known as the haem domain of soluble guanylate
cyclase (sGC), a mammalian nitric oxide receptor
Accepted 7 June, 2010. *For correspondence. E-mail marletta@
berkeley.edu; Tel. (+1) 510 666 2763; Fax (+1) 510 666 2765.
© 2010 Blackwell Publishing Ltd
(Derbyshire and Marletta, 2009). However, the recent
proliferation of sequenced prokaryotic genomes has
revealed a widespread distribution of H-NOX domains in
bacteria (Iyer et al., 2003; Karow et al., 2004; Boon
et al., 2006). All bacterial H-NOX domains expressed
and characterized thus far are relatively stable in the
ferrous oxidation state, bind diatomic gases, and are
found adjacent to or fused to signalling proteins (Iyer
et al., 2003; Karow et al., 2004; Boon et al., 2006).
These bacterial H-NOX proteins fall into at least two distinct classes based on their ligand binding characteristics and genomic context. In obligate-anaerobes, H-NOX
domains are fused to methyl accepting chemotaxis proteins. In facultative aerobic bacteria, H-NOX domains
are stand-alone proteins and are typically found adjacent
to histidine kinases or proteins involved in c-di-GMP
metabolism, such as those with GGDEF, EAL or
HD-GYP domains (Iyer et al., 2003). Biochemical characterization of prokaryotic H-NOXs has shown that the
H-NOX domains from obligate anaerobes are able to
bind O2, a property conferred upon them by the presence of a distal pocket tyrosine (Karow et al., 2004;
Boon et al., 2006). H-NOX proteins from facultative
aerobes lack this tyrosine and do not bind O2 (Karow
et al., 2004; Boon et al., 2005; 2006).
To date, all characterized H-NOX domains bind NO
tightly with a koff of ~10-3 s-1 (Boon et al., 2006). Assuming
a diffusion limited on-rate (kon) of ~108 M-1 s-1, which has
been observed for sGC and is likely to be conserved in the
prokaryotic H-NOX proteins (Stone and Marletta, 1996;
Zhao et al., 1999), the H-NOX family binds NO with picomolar affinity (Kd ~ 10 ¥ 10-12 M). Such tight binding suggests a potential biological function for the H-NOX as an
NO receptor. If these H-NOX domains do indeed function
as NO receptors, then they are tuned to sense concentrations several orders of magnitude below other characterized prokaryotic NO receptors, such as NorR, which
has a nanomolar Kd for NO (D’Autreaux et al., 2008) or
NsrR, which regulates gene transcription in response to
low micromolar levels of NO (Heurlier et al., 2008). Furthermore, the ability of the H-NOX proteins from facultative aerobes to not bind O2 distinguishes them from other
bacterial diatomic gas sensing haem proteins that have
this function (Tuckerman et al., 2009; Wan et al., 2009).
Thus, a reasonable hypothesis is that prokaryotic H-NOX
H-NOX regulation of biofilms 931
proteins from facultative aerobes are specific, high-affinity
NO receptors that sense picomolar levels of NO and
regulate cellular processes. H-NOX-mediated NO
sensing pathways may allow bacterial cells to adjust or
prepare for aspects of the environment that are signified
by low concentrations of NO. A hypothetical analogy may
be drawn to the NO signalling paradigm of mammalian
vascular nitric oxide signalling; a NO generating endothelial or neuronal cell produces a non-toxic, picomolar concentration of NO, which is sensed by soluble guanylate
cyclase in smooth muscle cells to cause vasodilation
(Derbyshire and Marletta, 2009).
Previous work suggests that the H-NOX proteins in
facultative aerobic bacteria are indeed involved in the
regulation of signalling pathways. In Shewanella oneidensis, Price et al. (2007) showed that the H-NOX inhibits the
activity of an adjacent histidine kinase in the NO-ligated
state, but not in the ferrous unligated state. A similar
mechanism of regulation has been observed for regulation of a phosphotransfer pathway by Hnox2 in Legionella
pneumophila (H. K. Carlson and M. A. Marletta, unpubl.
results).
In Vibrio fischeri, the H-NOX regulates a phosphotransfer pathway that modulates the colonization efficiency of
the bacteria in their consortial symbiotic association with
the squid host, Euprymna scolopes. H-NOX mutant
strains of V. fischeri hyper-colonize squid hosts 10-fold
better than wild-type bacteria. This striking result may be
explained by the upregulation of iron uptake genes in the
H-NOX mutants, which gives the mutants access to parts
of the squid crypts that are iron-limited and limit the growth
of wild-type bacteria (Wang et al., 2010). It may be that
the V. fischeri H-NOX represses iron-uptake genes
because the combination of NO and iron is toxic to the
bacterium. In this way, the H-NOX responds to low levels
of NO to prime the bacterial cell for higher, more toxic
levels of NO.
In this report, we have focused on L. pneumophila and
utilized phenotypic experiments to elucidate the role of
H-NOX proteins in this organism. As L. pneumophila has
two genes coding for H-NOX proteins, it provides a
unique opportunity to study the roles of H-NOX proteins
in regulating different types of signalling pathways
(Fig. 1). Both H-NOX proteins in L. pneumophila do not
bind O2, but do bind NO (Boon et al., 2006 and Fig. 1A).
One of the L. pneumophila H-NOX genes, lpg2459
(hnox2), is adjacent to a histidine kinase, lpg2458, and a
CheY-like response regulator, lpg2457. The other H-NOX
gene, lpg1056 (hnox1), is adjacent to a GGDEF-EAL
protein, lpg1057 (Fig. 1B). In the absence of clear phenotypic data, it is difficult to assign a function to CheYlike response regulators (Jenal and Galperin, 2009).
These response regulator proteins are known to affect
protein localization, function as phosphate sinks, or
control chemotactic or chemokinetic responses (Jenal
and Galperin, 2009). GGDEF-EAL proteins have a more
clearly defined role. These proteins are involved in the
metabolism of the bacterial second messenger, c-diGMP, regulating biofilm formation in a number of bacterial systems (Hengge, 2009). There may be spatial,
temporal or functional distribution within the family of
GGDEF-EAL proteins, but the processes that they regulate generally lead to an effect on the transition between
the biofilm-associated state of a bacterial cell and the
planktonic or virulent state (Tamayo et al., 2007; Hengge,
2009).
From a public health perspective, the mixed species
biofilms that form in anthropogenic water systems are well
known and important environmental reservoirs for L.
pneumophila growth (Declerck et al., 2009), and some
recent studies have provided insights into biofilm formation by L. pneumophila. It has been shown that L. pneumophila biofilm formation is more robust in rich media
compared with minimal media (Mampel et al., 2006), and
adheres to surfaces at 25°C better than at 37°C. At higher
temperatures, L. pneumophila forms longer mycelial matlike filaments in static cultures (Piao et al., 2006). In the
environment, L. pneumophila is found in biofilms in association with other bacteria, as well as protozoa. It has
been demonstrated that these interactions with other bacteria and protozoa facilitate the persistence of L. pneumophila in mixed-species biofilms (Mampel et al., 2006).
Thus far, the only genes clearly implicated in biofilm formation by L. pneumophila are fliA (Mampel et al., 2006)
and the tatB/C genes (De Buck et al., 2005). Genes
involved in c-di-GMP metabolism have not been characterized in L. pneumophila.
In this paper, we present evidence that an unmarked,
in-frame deletion of the L. pneumophila hnox1 gene
results in a hyper-biofilm phenotype, but that neither
hnox1 nor hnox2 deletions affect the virulence of L.
pneumophila in either amoebae or mouse macrophage
infections. Results from the biofilm assays lead us to
propose that the H-NOX proteins are likely to be NO
bound due to the presence of low nanomolar concentrations of NO present in the rich media used for
L. pneumophila growth. In addition, we show that
overexpression in L. pneumophila of Vca0956, a wellstudied diguanylate cyclase, and overexpression of
Lpg1057 both result in hyper-biofilm phenotypes. We
also demonstrate that the Lpg1057 protein has diguanylate cyclase activity in vitro that is inhibited by the presence of the H-NOX in the NO-ligated state. Finally, we
confirm the regulation of Lpg1057 by Hnox1 in vivo
through deletion of lpg1057 in the Dhnox1 background,
and show that the double mutant Dhnox1 Dlpg1057
displays wild-type levels of biofilm formation. Taken
together, these results suggest a role for Hnox1 as a
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
932 H. K. Carlson, R. E. Vance and M. A. Marletta !
Fig. 1. A. Electronic absorption spectra of the Fe(II) and Fe(II)-NO complexes for the two H-NOX proteins from L. pneumophila, showing the
Soret and a/b region. Haem concentration was ~5 mm.
B. The predicted H-NOX operons from L. pneumophila. The gene coding for Hnox1 is found adjacent to a diguanylate cyclase gene. The gene
coding for Hnox2 is found adjacent to a histidine kinase and single domain response regulator.
sensitive switch, responsive to picomolar levels of NO,
that regulates c-di-GMP metabolism and biofilm formation in L. pneumophila.
duced the same amount of pyomelanin pigment in late
post-exponential phase (Fig. S1).
Results
Dhnox1 strains display a hyper-biofilm phenotype in
BYE media
H-NOX proteins are not required for growth in rich
media (BYE), mouse macrophages or Acanthamoeba
castellanii
In order to determine the functional role of the H-NOX
proteins in L. pneumophila, deletions of both hnox1
(lpg1056) and hnox2 (lpg2459) as well as the double
mutant, Dhnox1 Dhnox2 were made as described in the
Experimental procedures. The growth kinetics of the
H-NOX mutant strains was identical to that of wild-type
LP02 in rich media (BYE) (Fig. S1), mouse macrophages
(Fig. S2) or Acanthamoeba castellanii (Fig. S3). All strains
became motile between an OD of 3.7 and 4.1, and pro-
A logical hypothesis for the function of the Hnox1 protein
is that it regulates the activity of Lpg1057, the GGDEFEAL protein adjacent to it in the L. pneumophila genome.
GGDEF-EAL proteins are implicated in c-di-GMP metabolism and the regulation of biofilm formation in a number of
prokaryotes (Tamayo et al., 2007; Hengge, 2009). In
general, bacterial cells with higher levels of c-di-GMP
produce more exopolysaccharides, adhere to solid surfaces more efficiently, form thicker biofilms, and are less
motile and less virulent (Tamayo et al., 2007; Hengge,
2009). However, in L. pneumophila, only three genes
have been reported to regulate biofilm formation, fliA
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
H-NOX regulation of biofilms 933
Fig. 2. Biofilm formation by LP02 and H-NOX mutants.
A. Optical density was measured at 600 nm (OD600) to quantify the growth of H-NOX mutant strains and LP02 in static culture in 96-well
polystyrene microtitre plates.
B. Biofilms were stained with crystal violet and resuspended in 95% ethanol before the absorbance at 600 nm (CV600) was measured to
quantify biofilm production for hnox1 mutant and LP02 in 96-well polystyrene microtitre plates.
C. Relative biofilm formation (relative CV600) at 5 days post inoculation for all strains in 96-well polystyrene microtitre plates.
D. L. pneumophila was grown without shaking for 5 days in 5 ml polystyrene culture tubes, and then adherent cells (biofilms) were stained
with crystal violet. *P < 0.05 for comparison of Dhnox1 and LP02. **P < 0.05 for comparison of Dhnox1 Dhnox2 and LP02.
(Mampel et al., 2006), tatB and tatC (De Buck et al.,
2005). We compared the growth and biofilm formation of
L. pneumophila wild-type and H-NOX mutant strains in
static cultures in BYE media. Initial observations of the
strains revealed that the Dhnox1 strains consistently
formed thicker biofilms. The hyper-biofilm phenotype was
quantified using crystal violet staining, and this assay was
used to show that strains lacking the hnox1 gene produced ~40% more biofilm mass than wild-type bacteria
(Fig. 2). Complementation of the Dhnox1 strains by
re-introduction of the Dhnox1 gene on a plasmid resulted
in a recovery of wild-type levels of biofilm formation upon
induction with IPTG (Fig. 3). These results clearly implicate hnox1 in the regulation of biofilm formation.
Furthermore, the H-NOX proteins are likely to be NO
bound due to the presence of millimolar concentrations of
NO3- and low nanomolar concentrations of NO in the rich
media used to grow L. pneumophila (Fig. S4A). The
chemistry responsible for the production of NO in BYE
involves ferric iron and L-cysteine that are commonly
added as supplements to support L. pneumophila growth
(Fig. S4B). Direct measurement of NO in growing static
cultures of L. pneumophila was not possible. However, it
is likely that the oxygen concentration in the static cultures
decreases greatly below the ring of growing biofilms at the
air–liquid interface. Sealed Hungate culture tubes were
used to mimic the microaerobic to anaerobic environment
present in static bacterial cultures, thereby providing a
closed environment to trap NO. Under these conditions
low nanomolar concentrations of NO were produced
(Fig. S4). Given that the H-NOX proteins likely have picomolar Kds for NO, the absence of the Hnox1 Fe(II)-NO
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
934 H. K. Carlson, R. E. Vance and M. A. Marletta !
Fig. 3. Complementation of hyper-biofilm
phenotype in hnox1 mutants. Biofilms were
stained with crystal violet, then resuspended
in 95% ethanol, and the absorbance at
600 nm (CV600) was measured. Relative CV600
staining is shown for wild-type and mutant
strains harbouring the pMMB206-GENT
vector, and the phnox1 vector with and
without IPTG induction. Biofilm formation was
quantified in 96-well polystyrene microtitre
plates (bars), and shown visually in 5 ml
polystyrene culture tubes (lower image) at 3
days post inoculation. *P < 0.05 for
comparison of Dhnox1 or Dhnox1
Dhnox2/pDhnox1 and LP02. **P < 0.05 for
comparison of uninduced and induced
Dhnox1/pDhnox1 or Dhnox1 Dhnox2/pDhnox1.
Induction of the phnox1 plasmid in LP02 had
no effect on biofilm formation (data not
shown).
species is likely responsible for the hyper-biofilm phenotype observed in the mutant strains.
Plasmid expression of diguanylate cyclase proteins in L.
pneumophila produces a hyper-biofilm phenotype, but
high levels of these proteins inhibits growth
To further investigate the role of c-di-GMP in biofilm formation in L. pneumophila, the vector pMMB206-Gent
(Hammer and Swanson, 1999; Molofsky and Swanson,
2003) was used to express two diguanylate cyclase proteins in L. pneumophila, Vca0956 (Tischler and Camilli,
2004) and Lpg1057. pMMB206-Gent has a Ptac promoter
and is known to have leaky expression when used in L.
pneumophila (Molofsky and Swanson, 2003). Therefore, it
was not surprising to find that both the pvca0956 and
plpg1057 strains displayed hyper-biofilm phenotypes in
the absence of IPTG (Fig. 4A). Induction with IPTG inhibited growth of the strains, suggesting that higher levels of
c-di-GMP are toxic or severely inhibitory on L. pneumophila growth (Fig. 4A). In shaken cultures, IPTG induction
of pvca0956 and plpg1057 produced more filamentous
cells in early post-exponential phase than wild-type
(Fig. 4B). Also, nucleotide extracts from cultures overexpressing pvca0956 or plpg1057 had higher concentrations of c-di-GMP than vector controls (Fig. S5).
Filamentous L. pneumophila are observed in late PE
phase, or in static cultures grown at 37°C in the form of
mycelial mat-like biofilms (Piao et al., 2006). These observations further support the role of higher c-di-GMP levels
in biofilm formation in L. pneumophila, and support the
hypothesis that Lpg1057 has dominant diguanylate
cyclase activity in vivo unless it is inhibited by the Hnox1
protein.
Lpg1057 is an active diguanylate cyclase and is
inhibited by Hnox1 Fe(II)-NO
To better understand the role of the Hnox1 protein in the
regulation of biofilm formation, Hnox1 and Lpg1057 proteins and Vca0956, an active GGDEF domain containing
protein from V. cholerae, were expressed and purified
(Tischler and Camilli, 2004; Thormann et al., 2006). We
found that Lpg1057 was an active diguanylate cyclase,
though over 100-fold less active than Vca0956 (Fig. 5).
The specific activity of the Lpg1057 protein preparation
was estimated to be 0.04 ! 0.02 nmol min-1 mg-1 based
on the quantification of the 24-hour time point, but this
value is likely to be a lower limit since the protein concentration had to be estimated (~50% pure). If Hnox1 is acting
as a NO responsive switch to regulate Lpg1057 in vivo,
then it should alter the production of c-di-GMP by the
Lpg1057 protein in vitro. Hnox1 Fe(II)-NO inhibited the
formation of c-di-GMP by Lpg1057, but Hnox1 Fe(II) unligated had no effect (Fig. 6). Unfortunately, due to the low
activity of the Lpg1057 protein, 24-hour reaction times
were required to see quantifiable c-di-GMP in these
assays. The low diguanylate cyclase activity of Lpg1057
may be due to the presence of a c-di-GMP phosphodiesterase, EAL, domain. However, phosphodiesterase
assays did not reveal any measurable phosphodiesterase
activity (data not shown). Furthermore, the Lpg1057 EAL
domain has non-conservative mutations in a conserved
active site loop that may render this domain catalytically
inactive (Rao et al., 2009). The low activity of Lpg1057
limits further conclusions. However, given the slow growth
rate of L. pneumophila and the potential for c-di-GMP
signalling pathways to be spatially or temporally located
(Hengge, 2009), this low activity may be sufficient to
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
H-NOX regulation of biofilms 935
Fig. 4. Effect of diguanylate cyclase protein expression on L. pneumophila biofilm formation. Growth of static cultures was monitored by
measuring the optical density at 600 nm (OD600). Biofilms were stained with crystal violet and resuspended in 95% ethanol before the
absorbance at 600 nm (CV600) was measured.
A. The relative CV600 and relative OD600 of static cultures in 96-well polystyrene microtitre plates at 5 days post-inoculation for LP02 harbouring
pMMB206-GENT pvca0956 or plpg1057.
B. 40¥ magnification images of post-exponential (OD600 = 4.0) L. pneumophila grown in liquid cultures with shaking and 100 mM IPTG
induction. LP02 harbouring the pMMB206-GENT vector forms coccoid, motile cells in post-exponential phase, but Vca0956 and Lpg1057
induce filamentous morphology. **P < 0.05 for CV600 comparison of LP02 harbouring the empty pMMB-206-GENT vector with LP02 harbouring
pvca0956 or plpg1057. *P < 0.05 for OD600 comparison of LP02 harbouring the empty pMMB-206-GENT vector with LP02 harbouring
pvca0956 or plpg1057.
Fig. 5. 2D TLC analysis of GTP turnover to
c-di-GMP by Vca0956 and Lpg1057. The top
row shows the crude mixture of products after
24 h at 25°C for no enzyme, Vca0956
(~1 mM) or Lpg1057 (~1 mM) incubated with
500 mM GTP and 10 mCi GTPa32P. Arrows
indicate the c-di-GMP spot. The bottom row
shows the same reactions after treatment with
CIP (calf intestine alkaline phosphatase).
c-di-GMP is resistant to CIP treatment and
remains, while other nucleotides are
degraded.
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
936 H. K. Carlson, R. E. Vance and M. A. Marletta !
Fig. 6. Effect of Hnox1 Fe(II) and Hnox1 Fe(II)-NO on c-di-GMP production by Lpg1057 in vitro.
A. 1D TLC analysis of GTP turnover to c-di-GMP by Lpg1057. Lpg1057 (~ 1 mM) in the presence of Hnox1 (50 mM) in the Fe(II)-NO or Fe(II)
unligated states was incubated in sealed anaerobic vials with 500 mM GTP and 10 mCi GTPa32P for 24 h at 25°C and then treated with CIP.
Arrows indicate the c-di-GMP spot. Spectral analysis of identical reactions without radioactive GTP revealed that less than 10% of the Fe(II)
protein was oxidized after 24 h in the sealed vials, and the Fe(II)-NO complex was stable.
B. Relative c-di-GMP production for the Lpg1057 protein alone or in the presence of the Hnox1 Fe(II)-NO or Hnox1 Fe(II). The intensity of the
c-di-GMP spots in the phosphorimage were quantified using ImageQuant software (Molecular Dynamics). The error bars represent average
values obtained from three independent experiments. *P < 0.05 for comparison of c-di-GMP production of Lpg1057 with Lpg1057 + Hnox1
Fe(II)NO.
explain the hyper-biofilm phenotype of the Dhnox1 strains.
Furthermore, the comparison with the highly active
Vca0956 may be misleading. Nucleotide extracts from L.
pneumophila strains overexpressing Vca0956 had higher
concentrations of c-di-GMP compared with strains overexpressing Lpg1057 (Fig. S5).
Dlpg1057 strains display wild-type biofilm growth
We hypothesized that the hyper-biofilm phenotype
observed in the Dhnox1 strains was due to a lack of
inhibition of the diguanylate cyclase activity of Lpg1057 by
the H-NOX. To gather support for this hypothesis, deletions of the lpg1057 gene in the Dhnox1 background were
made. We found, as expected, that the deletion of the
lpg1057 gene reversed the hyper-biofilm phenotype of the
hnox1 mutants (Fig. 7). This finding clearly demonstrates
that Lpg1057 is part of the Hnox1 signalling pathway.
Discussion
Legionella H-NOX proteins are highly sensitive, specific
NO sensors
Fig. 7. Biofilm formation by LP02, Dhnox1, and Dhnox1 Dlpg1057
strains. Biofilms were stained with crystal violet, then resuspended
in 95% ethanol, and the absorbance at 600 nm (CV600) was
measured. Relative CV600 was quantified for 96-well polystyrene
microtitre plates (top), and shown visually for crystal violet stained
5 ml polystyrene culture tubes for wild-type LP02 and mutants in
the hnox1 signalling pathway at 5 days post-inoculation (bottom).
*P < 0.05 for comparison of LP02 with Dhnox1. **P < 0.05 for
comparison of Dhnox1 with Dhnox1 Dlpg1057.
Previous studies have found that the H-NOX domains
from facultative aerobic bacteria do not form an O2
complex, including the two H-NOX proteins from L. pneu© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
H-NOX regulation of biofilms 937
mophila (Karow et al., 2004; Boon et al., 2006). In contrast, the H-NOX domains from obligate anaerobes form
stable and tight complexes with O2, a property conferred
upon them by the presence of a distal pocket tyrosine
(Boon et al., 2005). The non-O2 binding H-NOX domains
are distinctive among haem proteins in their ability to bind
NO with high affinity in the presence of much higher
concentrations of oxygen. The L. pneumophila Hnox1
protein involved in the biofilm phenotype reported
here has a koff of ~10-3 s-1 at 20°C (Boon et al., 2006),
and assuming a nearly diffusion limited kon of
~4.5 ¥ 108 M-1 s-1 at 20°C (Stone and Marletta, 1996;
Zhao et al., 1999), the Kd would be in the picomolar range.
This high affinity Kd suggests that the H-NOX is likely to be
NO bound in the rich media used for L. pneumophila. The
media contains millimolar concentrations of NO3-, and low
nanomolar concentrations of NO produced by the iron and
L-cysteine dependent, heat-catalysed, inorganic decomposition of the NO3- (Fig. S4). Therefore, any phenotypes
observed in the mutant strains grown in rich media are
likely due to the absence of the H-NOX Fe(II)-NO species.
L. pneumophila H-NOX proteins, like all H-NOX domains,
also bind CO and can be oxidized. However, in previous
studies, CO-bound H-NOX domains were found to have a
much weaker effect on the regulation of other protein
domains compared with NO bound H-NOX (Price et al.,
2007; Derbyshire and Marletta, 2009). Furthermore, L.
pneumophila does not possess the CO dehydrogenase
that has been shown to be responsible for CO production
in other bacteria (Roberts et al., 2004). Although the
H-NOX domain from S. oneidensis inhibits its associated
kinase in the Fe(III) state, it is likely that H-NOX domains
are in the Fe(II) state in the reducing environment of the
bacterial cell (Price et al., 2007). It is more likely that
proteins such as OxyR, which is present in L. pneumophila, are responsible for sensing oxidative stress (Christman et al., 1989; LeBlanc et al., 2008).
Legionella pneumophila could be exposed to NO in a
number of ways. Both eukaryotic and bacterial cells
produce NO through the action of nitric oxide synthase
(NOS) (Shatalin et al., 2008; Agapie et al., 2009;
Sudhamsu and Crane, 2009). Denitrifying bacteria also
produce NO as an intermediate of this process (Corker
and Poole, 2003; Bedmar et al., 2005; Rodionov et al.,
2005), but NO3– can also be chemically reduced by Fe(II)
in the presence of catalytic amounts of other transition
metals, such as Cu(II) (Ottley et al., 1997). The chemical
reduction of NO3– is rapid at high temperatures and pressures, and is an important nitrogen sink and likely NO
source in extreme environments, such as hydrothermal
vents or the early earth (Brandes et al., 1998). In fact, it
has been proposed that NO was the first biological terminal electron sink to emerge on the earth before the
appearance of molecular oxygen (Ducluzeau et al.,
2009). The photolysis of NO3- to NO2- to NO is another an
important abiotic source of NO, and this process is at least
partly responsible for the halo of low nanomolar NO levels
that encircles the planet at the bottom of the photic zone
in the oceans (Zafiriou et al., 1980). Bacteria with H-NOX
proteins are found in all of these environments, and the
H-NOX may be regulating biofilm formation in response to
NO in these environments.
Deletions of L. pneumophila hnox genes do not affect
growth in liquid culture or phagocytes
A well-studied part of the L. pneumophila cell cycle is
the transition from the replicative phase during exponential growth in liquid culture or phagocytes to the transmissive phase during post-exponential, stationary growth.
This transition is regulated by a ppGpp alarmone
response to starvation conditions and involves a number
of well-studied regulatory genes such as csrA, letA and
fliA (Byrne and Swanson, 1998; Hammer and Swanson,
1999; Molofsky and Swanson, 2004). If the H-NOX were
regulating any of these genes, it would likely have
affected the virulence of the mutant strains, but we did not
observe virulence phenotypes for the H-NOX mutants
(Fig. S2, Fig. S3).
A few studies have focused on the potential role for NO
in L. pneumophila infections. As an intracellular pathogen,
L. pneumophila potentially encounters toxic levels of
immune-derived NO since iNOS is induced in L. pneumophila infected macrophages (Heath et al., 1996).
However, the addition of iNOS inhibitors to permissive A/J
macrophages infected with L. pneumophila had little
effect on the intracellular growth of the bacteria (Gebran
et al., 1994; Yamamoto et al., 1996). In L. pneumophila
infected A/J mice, however, the NOS inhibitor, N-methylL-arginine, diminished the ability of the mice to clear the
bacteria (Brieland et al., 1995). Hence, although NO may
not be directly involved in killing intracellular bacteria, it
may play a role in the immune response to Legionella. NO
is clearly toxic to L. pneumophila, and it would be interesting to know if L. pneumophila strains with deletions in
the nsrR gene (Rodionov et al., 2005) or the genes coding
for NO reductases (Rodionov et al., 2005) are deficient in
replication within macrophages or amoebae.
L. pneumophila H-NOX proteins are NO sensitive
switches that regulate biofilm formation
Regulation of biofilm attachment, detachment and growth
by changes in the concentration of diatomic gases such
as NO or O2 has been observed in several bacteria (Thormann et al., 2005; Barraud et al., 2006; 2009). c-di-GMP
metabolism is well established as a regulator of biofilm
formation (Tamayo et al., 2007; Hengge, 2009), and it has
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
938 H. K. Carlson, R. E. Vance and M. A. Marletta !
been proposed that protein domains involved in sensing
diatomic gases act as in vivo switches to regulate
domains involved in c-di-GMP metabolism (Tuckerman
et al., 2009; Wan et al., 2009).
Microarrays show that the L. pneumophila H-NOX
genes are upregulated alongside the other proteins in
their predicted operons in transmissive phase, as are
many other proteins involved in virulence (Bruggemann
et al., 2006). Also upregulated in transmissive phase are
proteins involved in c-di-GMP metabolism (Bruggemann
et al., 2006). These results suggest that more developmental checkpoints guiding the decision to form a
biofilm or to infect phagocytes may exist in transmissive phase, rather than replicative. In support of this
hypothesis, several studies show that transmissive
phase L. pneumophila in static cultures adhere to solid
surfaces in the initial phase of biofilm attachment better
than replicative phase L. pneumophila (Hindre et al.,
2008).
Some haem-containing diatomic gas sensors have
been shown to regulate diguanylate cyclase and phosphodiesterase domains, but these proteins bind both NO
and O2 and their activity is affected by both ligands (Tuckerman et al., 2009; Wan et al., 2009). However, the L.
pneumophila H-NOX protein does not bind O2; thus, NO is
the most likely physiological ligand. Hence, this study is
the first to describe a biofilm phenotype that implicates
both a diatomic gas sensor that does not bind O2, the
Hnox1, and a protein involved in c-di-GMP metabolism,
Lpg1057.
Relatively few L. pneumophila mutant strains are
reported to have a biofilm phenotype. To date, only fliA
mutants and tatB/C mutants have been shown to have
defects in biofilm formation (De Buck et al., 2005; Hindre
et al., 2008). Microarrays suggest that genes involved in
the response to oxidative stress and iron uptake are
upregulated in biofilms (Hindre et al., 2008), and supplementation of static cultures with excess iron has been
shown to inhibit biofilm formation (Hindre et al., 2008). It is
highly likely that mutations in many of the genes involved
in the metabolism of c-di-GMP will display biofilm phenotypes, but no other studies have analysed the role of
these genes in L. pneumophila.
The results presented in this paper suggest that the
Hnox1 protein regulates biofilm formation in response to
NO by altering the c-di-GMP production activity of the
Lpg1057 protein (Fig. 8). This is the first report of biofilm
formation being regulated by c-di-GMP in L. pneumophila.
Future studies will focus on the role of H-NOX proteins in
the regulation of biofilm formation in other bacteria, and
how the H-NOX domains may work in concert with other
diatomic gas sensor proteins to allow L. pneumophila to
persist and survive in environments with varying concentrations of NO.
Fig. 8. A model for the hnox1 pathway in L. pneumophila. The
Hnox1 protein inhibits the diguanylate cyclase activity of Lpg1057
in response to NO, leading to lower c-di-GMP levels and less
biofilm formation. Dhnox1 strains produce more biofilms because
c-di-GMP production by Lpg1057 protein is no longer inhibited by
Hnox1.
Experimental procedures
Bacterial strains, culture conditions, and reagents
LP02 is a streptomycin-resistant thymidine auxotroph derived
from LP01. All mutant strains were made in the LP02
background. L. pneumophila was grown on ACES buffered
yeast extract (BYE) charcoal agar plates (BCYE) supplemented with 0.1 g l-1 thymidine, 0.4 g l-1 L-cysteine and
0.135 g l-1 ferric nitrate. Liquid cultures in BYE with supplements were inoculated with patches of L. pneumophila
strains. Antibiotics were added to BCYE plates or liquid BYE,
when needed at the following concentrations: streptomycin
100 mg ml-1, kanamycin, 25 mg ml-1, and gentamicin,
10 mg ml-1. The optical density at 600 nm (OD600) of cultures
was measured by diluting the cultures 10-fold in BYE in 1 cm
plastic cuvettes and reading the OD600 with an Ultraspec 100
cell density reader (Amersham Biosciences). An OD600 of 3
represents approximately 109 bacteria·per millilitre.
Mutant strain construction
Unmarked in-frame deletions in the genes coding for the two
H-NOX proteins and the GGDEF-EAL protein were constructed using previously published methods (Merriam et al.,
1997; Ren et al., 2006). Briefly, the flanking regions of the
genes to be deleted were cloned into the vector pSR47S and
transformed into DH5a lpir E. coli. For matings, patches of L.
pneumophila were mixed with the mating strain containing
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
H-NOX regulation of biofilms 939
Table 1. Strains, plasmids and primers used in this study.
Strains and plasmids
Bacterial strains
E. coli
DH5a lpir
2174
BL21(DE3) pLysS
L. pneumophila
LP02
Dhnox1
Dhnox2
Dhnox1 Dhnox2
Dhnox1 Dlpg1057
LP02/ pJB908
Dhnox1/ pJB908
Dhnox2/ pJB908
Dhnox1 Dhnox2/ pJB908
LP02/pMMB206-GENT
LP02/pvca0956
LP02/plpg1057
Dhnox1/phnox1
Plasmids
pSR47S
pSRhnox1
pSRhnox2
pSRlpg1057
pJB908
pMMB206-GENT
plpg1057
pvca0956
phnox1
pET20-b(+)
pEThnox1
pETlpg1057
pAC1758
Relevant characteristics
Sources or references
Mating strain
Helper strain
Expression strain
Merriam et al. (1997); Ren et al. (2006)
Merriam et al. (1997); Ren et al. (2006)
Invitrogen
Wild-type L. pneumophila, thymidine auxotroph Smr
LP02 Dhnox1 Smr
LP02 Dhnox2 Smr
LP02 Dhnox1 Dhnox2 Smr
LP02 Dhnox1 Dlpg1057 Smr
LP02 pJB908 Smr Ampr
LP02 Dhnox1 pJB908Smr Ampr
LP02 Dhnox2 pJB908Smr Ampr
LP02 Dhnox1 Dhnox2 pJB908 Smr Ampr
LP02 pMMB206-GENT Smr Gentr
LP02 pvca0956 Smr Gentr
LP02 plpg1057 Smr Gentr
Dhnox1 phnox1 Smr Gentr
This
This
This
This
This
This
This
This
This
This
This
This
sacB Kanr
psR47S::Dhnox1 sacB Kanr
psR47S::Dhnox2 sacB Kanr
psR47S::Dlpg1057 sacB Kanr
thymidylate synthetase Ampr
pTac promoter Gentr
pMMB206-GENT::lpg1057
pMMB206-GENT::vca0956
pMMB206-GENT::hnox1
T7 promoter E. coli expression vector Ampr
pET20-b(+)::Hnox1-His6 Ampr
pET20-b(+)::Lpg1057-His6 Ampr
pBAD33::Vca0956-His6, Cmr
Merriam et al. (1997); Ren et al. (2006)
This work
This work
This work
Gift from Professor Ralph Isberg (Tufts University)
Hammer and Swanson (1999)
This work
This work
This work
Invitrogen
This work
This work
Gift from Professor Andrew Camilli (Tufts University)
the knock-out vector and the helper strain 2174. Matings
were spread on BCYE plates containing 100 mg ml-1 streptomycin and 25 mg ml-1 kanamycin to select for the first crossover event. The second cross-over event was selected for on
BCYE plates containing 100 mg ml-1 streptomycin and 6%
sucrose. Colonies that did not grow when patched onto
BCYE plates containing 25 mg ml-1 kanamycin were deletion
candidates. These candidates were confirmed by PCR and
sequenced with primers flanking the gene of interest to verify
the presence of clean in-frame deletions. Multiple isolates of
each mutant strain were tested to confirm the phenotypes
reported in this paper.
Strains and plasmids
Strains and plasmids are listed in Table 1. Oligonucleotide
primers are listed in Table S1. A detailed description of
cloning techniques is described in Appendix S1.
L. pneumophila biofilm formation assay
Legionella pneumophila biofilms were generated as
described previously with slight modifications (Mampel et al.,
2006; Hindre et al., 2008). Overnight cultures of L. pneumophila in exponential phase (OD600 = 3) were diluted 10¥ into a
work
work
work
work
work
work
work
work
work
work
work
work
final volume of 200 ml fresh BYE in 96-well polystyrene
microtitre plates (Costar). Biofilm formation was induced by
placing the microtitre plates containing bacteria into a 30°C
incubator and growing static cultures for between 1 and 7
days. Alternatively, 5 ml polystyrene culture tubes were
inoculated with 10¥ dilutions of OD600 = 3 bacteria and grown
for between 1 and 7 days at 30°C.
Quantification of L. pneumophila biofilm formation
Legionella pneumophila biofilm formation was quantified
essentially as described previously (Mampel et al., 2006;
Hindre et al., 2008). The optical density of static cultures in
the 96-well microplates was determined by reading OD600 on
a SpectraMax M2 optical plate reader (Molecular Dynamics).
At the indicated timepoints, the planktonic bacteria were
removed and the plate was washed twice with 200 ml BYE per
well. Crystal violet (0.3%; 200 ml) was then added to the wells
for 10 min to stain the biofilms. The excess crystal violet was
removed and the wells were washed 3 times with 200 ml
water. To quantify biofilm formation, 200 ml of ethanol was
added to the wells to resuspend the biofilms and the OD600 of
each well was read on the plate reader. Two-tailed Student’s
t-tests were used to analyse the data. P-values of < 0.05
were considered significant.
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
940 H. K. Carlson, R. E. Vance and M. A. Marletta !
Expression and purification of Hnox1, Lpg1057, and
Vca0956 proteins
Legionella pneumophila Hnox1 (Lpg1056) and GGDEF-EAL
protein (Lpg1057) were cloned out of genomic DNA, and
ligated into the pET-20b(+) expression vector (Invitrogen) cut
with NcoI and XbaI restriction enzymes (NEB), and then
transformed into E. coli DH5a. Positive transformants of all
constructs were screened for on LB plates containing
100 mg ml-1 ampicillin, and the DNA sequences were confirmed by sequencing (Elim Biopharmaceuticals). All proteins
were expressed as follows: E. coli BL21(DE3)pLysS cells
containing the appropriate plasmid were grown at 37°C to an
OD600 of 0.6–0.9, induced with 10 mM IPTG, and grown for
16–18 h at 25°C. Cells were harvested by centrifugation at
7000 r.p.m. (6370 g) for 15 min in an Avanti J20I centrifuge
with a JLA 8.1 rotor (Beckman), resuspended in lysis buffer
[100 mM sodium phosphate, 250 mM NaCl, 5% glycerol,
20 mM imidazole, 1 mM Pefabloc (Roche), pH 7.9] and lysed
with a high-pressure homogenizer (Avestin). Lysate was
clarified by centrifugation at 42 000 r.p.m. (200 000 g) for 1 h
in an Optima Xl-100K ultracentrifuge with a Ti-45 rotor
(Beckman) prior to loading on Ni-NTA agarose resin
(Qiagen). The Ni-NTA resin was washed with lysis buffer and
eluted with elution buffer (250 mM imidazole in 100 mM
sodium phosphate, 250 mM NaCl, 5% glycerol, pH 7.9). For
Hnox1, gel filtration chromatography with a S200 26/60
HiLoad Resin column (Pharmacia Biotech) connected to a
Biologic HR FPLC was used for further purification. The gel
filtration column was equilibrated and run in 50 mM TEA,
50 mM NaCl, 5% glycerol, pH 7.5 buffer. Hnox1 was > 95%
pure for assays, as determined by SDS-PAGE and Coomassie staining. The diguanylate cyclases used in the assays
were approximately 50% pure as assessed by Coomassie
staining. Protein concentrations were determined by the
Bradford method.
Spectral analysis of Hnox1 and preparation of Fe(II)
and Fe(II)-NO
Purified H-NOX protein was desalted into the spectral buffer
(50 mM TEA, 50 mM NaCl, pH 7.5) using a PD-10 desalting
column (GE Healthcare) in an anaerobic glove bag. The
protein was then oxidized with 20-fold molar excess of potassium ferricyanide, and desalted to give the oxidized protein.
Oxidized protein was reduced with a 50-fold excess of
sodium dithionite, and desalted to give the ferrous unligated
species, H-NOX Fe(II)-NO was added by providing a 10-fold
excess of DEA-NONOate (Cayman Chemical) to the reduced
protein preparation, and the protein was desalted to give the
ferrous nitrosyl complex, H-NOX Fe(II)-NO. The H-NOX
spectra were recorded on a Cary 3E spectrophotometer at
20°C.
In vitro diguanylate cyclase assays
The activity assay for the GGDEF-EAL protein adjacent to
the H-NOX for the formation or degradation of c-di-GMP
was done as reported (Tamayo et al., 2005) with some
modifications. Due to difficulties in obtaining pure Lpg1057,
protein that was approximately 50% pure was used in this
assay. For diguanylate cyclase assays, Vca0956 or Lpg1057
(~1 mM) were incubated with 500 mM GTP and 10 mCi
GTPa32P for 24 h at 25°C. Products were treated with calf
intestine alkaline phosphatase (CIP) (New England Biolabs)
for 2 h. Nucleotides were separated from protein with a 10K
MWCO spin filter and spotted onto PEI-cellulose TLC plates.
For 2D TLCs, the plates were developed first with 0.2 M
NH4HCO3, pH 7.5 and then with 1.5 M KH2PO4, pH 4. For
1D TLCs, the plates were developed with 1.5 M KH2PO4, pH
4. Developed TLC plates were exposed to a phosphorimager plates (Molecular Dynamics) and imaged with a
Typhoon (Molecular Dynamics), and the intensity of the
spots was quantified using ImageQuant software (Molecular
Dynamics).
For assays with H-NOX proteins, ~1 mM Lpg1057 and
50 mM Hnox1 were mixed in an anaerobic glove bag (Coy) in
sealed vials. A small volume of aerobic GTP was added with
a gas-tight syringe (Hamilton) to give a final concentration,
500 mM GTP and 10 mCi GTPa32P, and timepoints were
removed with the gas-tight syringe between 0 and 24 h at
25°C. Control vials in which no radioactive GTP was added
were also prepared, and spectral analysis of the Hnox1 with
a Cary 3E spectrophotometer revealed that less than 10% of
the total Hnox1 Fe(II) was oxidized after overnight reactions,
and the Hnox1 Fe(II)-NO complex was completely intact.
Acknowledgements
We thank Drs Michele Swanson (University of Michigan) and
Jean-Francois Dubuisson (University of Michigan) for helpful
advice, and for early experiments related to this project. We
are also grateful to members of the Vance and Marletta labs
for helpful advice and critical comments on this manuscript.
This work was supported by NIH grants GM070671 (M.A.M.)
and AI075039 and AI080749 (R.E.V.) and Investigator
Awards from the Burroughs Wellcome Fund and the Cancer
Research Institute (R.E.V.).
References
Agapie, T., Suseno, S., Woodward, J.J., Stoll, S., Britt, R.D.,
and Marletta, M.A. (2009) NO formation by a catalytically
self-sufficient bacterial nitric oxide synthase from Sorangium cellulosum. Proc Natl Acad Sci USA 106: 16221–
16226.
Barraud, N., Hassett, D.J., Hwang, S.H., Rice, S.A., Kjelleberg, S., and Webb, J.S. (2006) Involvement of nitric oxide
in biofilm dispersal of Pseudomonas aeruginosa.
J Bacteriol 188: 7344–7353.
Barraud, N., Schleheck, D., Klebensberger, J., Webb, J.S.,
Hassett, D.J., Rice, S.A., and Kjelleberg, S. (2009) Nitric
oxide signaling in Pseudomonas aeruginosa biofilms mediates phosphodiesterase activity, decreased cyclic di-GMP
levels, and enhanced dispersal. J Bacteriol 191: 7333–
7342.
Bedmar, E.J., Robles, E.F., and Delgado, M.J. (2005) The
complete denitrification pathway of the symbiotic, nitrogenfixing bacterium Bradyrhizobium japonicum. Biochem Soc
Trans 33: 141–144.
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
H-NOX regulation of biofilms 941
Boon, E.M., Huang, S.H., and Marletta, M.A. (2005) A
molecular basis for NO selectivity in soluble guanylate
cyclase. Nat Chem Biol 1: 53–59.
Boon, E.M., Davis, J.H., Tran, R., Karow, D.S., Huang, S.H.,
Pan, D., et al. (2006) Nitric oxide binding to prokaryotic
homologs of the soluble guanylate cyclase beta1 H-NOX
domain. J Biol Chem 281: 21892–21902.
Brandes, J.A., Boctor, N.Z., Cody, G.D., Cooper, B.A.,
Hazen, R.M., and Yoder, H.S., Jr (1998) Abiotic nitrogen
reduction on the early Earth. Nature 395: 365–367.
Brieland, J.K., Remick, D.G., Freeman, P.T., Hurley, M.C.,
Fantone, J.C., and Engleberg, N.C. (1995) In vivo regulation of replicative Legionella pneumophila lung infection by
endogenous tumor necrosis factor alpha and nitric oxide.
Infect Immun 63: 3253–3258.
Bruggemann, H., Hagman, A., Jules, M., Sismeiro, O., Dillies,
M.A., Gouyette, C., et al. (2006) Virulence strategies for
infecting phagocytes deduced from the in vivo transcriptional program of Legionella pneumophila. Cell Microbiol 8:
1228–1240.
Byrne, B., and Swanson, M.S. (1998) Expression of
Legionella pneumophila virulence traits in response to
growth conditions. Infect Immun 66: 3029–3034.
Christman, M.F., Storz, G., and Ames, B.N. (1989) OxyR, a
positive regulator of hydrogen peroxide-inducible genes in
Escherichia coli and Salmonella typhimurium, is homologous to a family of bacterial regulatory proteins. Proc Natl
Acad Sci USA 86: 3484–3488.
Corker, H., and Poole, R.K. (2003) Nitric oxide formation by
Escherichia coli. Dependence on nitrite reductase, the
NO-sensing regulator Fnr, and flavohemoglobin Hmp.
J Biol Chem 278: 31584–31592.
D’Autreaux, B., Tucker, N., Spiro, S., and Dixon, R. (2008)
Characterization of the nitric oxide-reactive transcriptional
activator NorR. Methods Enzymol 437: 235–251.
De Buck, E., Maes, L., Meyen, E., Van Mellaert, L., Geukens
N., Anne, J., and Lammertyn, E. (2005) Legionella pneumophila Philadelphia-1 tatB and tatC affect intracellular
replication and biofilm formation. Biochem Biophys Res
Commun 331: 1413–1420.
Declerck, P., Behets, J., Margineanu, A., van Hoef, V., De
Keersmaecker, B., and Ollevier, F. (2009) Replication of
Legionella pneumophila in biofilms of water distribution
pipes. Microbiol Res 164: 593–603.
Derbyshire, E.R., and Marletta, M.A. (2009) Biochemistry of
soluble guanylate cyclase. Handb Exp Pharmacol 17–31.
Ducluzeau, A.L., van Lis, R., Duval, S., Schoepp-Cothenet,
B., Russell, M.J., and Nitschke, W. (2009) Was nitric oxide
the first deep electron sink? Trends Biochem Sci 34: 9–15.
Gebran, S.J., Yamamoto, Y., Newton, C., Klein, T.W., and
Friedman, H. (1994) Inhibition of Legionella pneumophila
growth by gamma interferon in permissive A/J mouse macrophages: role of reactive oxygen species, nitric oxide,
tryptophan, and iron(III). Infect Immun 62: 3197–3205.
Hammer, B.K., and Swanson, M.S. (1999) Co-ordination of
Legionella pneumophila virulence with entry into stationary
phase by ppGpp. Mol Microbiol 33: 721–731.
Heath, L., Chrisp, C., Huffnagle, G., LeGendre, M., Osawa,
Y., Hurley, M., et al. (1996) Effector mechanisms responsible for gamma interferon-mediated host resistance to
Legionella pneumophila lung infection: the role of endog-
enous nitric oxide differs in susceptible and resistant
murine hosts. Infect Immun 64: 5151–5160.
Hengge, R. (2009) Principles of c-di-GMP signalling in
bacteria. Nat Rev Microbiol 7: 263–273.
Heurlier, K., Thomson, M.J., Aziz, N., and Moir, J.W. (2008)
The nitric oxide (NO)-sensing repressor NsrR of Neisseria
meningitidis has a compact regulon of genes involved in
NO synthesis and detoxification. J Bacteriol 190: 2488–
2495.
Hindre, T., Bruggemann, H., Buchrieser, C., and Hechard, Y.
(2008) Transcriptional profiling of Legionella pneumophila
biofilm cells and the influence of iron on biofilm formation.
Microbiology 154: 30–41.
Iyer, L.M., Anantharaman, V., and Aravind, L. (2003) Ancient
conserved domains shared by animal soluble guanylyl
cyclases and bacterial signaling proteins. BMC Genomics
4: 5.
Jenal, U., and Galperin, M.Y. (2009) Single domain response
regulators: molecular switches with emerging roles in cell
organization and dynamics. Curr Opin Microbiol 12: 152–
160.
Karow, D.S., Pan, D., Tran, R., Pellicena, P., Presley, A.,
Mathies, R.A., and Marletta, M.A. (2004) Spectroscopic
characterization of the soluble guanylate cyclase-like heme
domains from Vibrio cholerae and Thermoanaerobacter
tengcongensis. Biochemistry 43: 10203–10211.
LeBlanc, J.J., Brassinga, A.K., Ewann, F., Davidson, R.J.,
and Hoffman, P.S. (2008) An ortholog of OxyR in
Legionella pneumophila is expressed postexponentially
and negatively regulates the alkyl hydroperoxide reductase
(ahpC2D) operon. J Bacteriol 190: 3444–3455.
Mampel, J., Spirig, T., Weber, S.S., Haagensen, J.A., Molin,
S., and Hilbi, H. (2006) Planktonic replication is essential
for biofilm formation by Legionella pneumophila in a
complex medium under static and dynamic flow conditions.
Appl Environ Microbiol 72: 2885–2895.
Merriam, J.J., Mathur, R., Maxfield-Boumil, R., and Isberg,
R.R. (1997) Analysis of the Legionella pneumophila fliI
gene: intracellular growth of a defined mutant defective for
flagellum biosynthesis. Infect Immun 65: 2497–2501.
Molofsky, A.B., and Swanson, M.S. (2003) Legionella pneumophila CsrA is a pivotal repressor of transmission traits
and activator of replication. Mol Microbiol 50: 445–461.
Molofsky, A.B., and Swanson, M.S. (2004) Differentiate to
thrive: lessons from the Legionella pneumophila life cycle.
Mol Microbiol 53: 29–40.
Ottley, C.J., Davison, W., and Edmunds, W.M. (1997) Chemical catalysis of nitrate reduction by iron(II). Geochimica Et
Cosmochimica Acta 61: 1819–1828.
Piao, Z., Sze, C.C., Barysheva, O., Iida, K., and Yoshida, S.
(2006) Temperature-regulated formation of mycelial matlike biofilms by Legionella pneumophila. Appl Environ
Microbiol 72: 1613–1622.
Price, M.S., Chao, L.Y., and Marletta, M.A. (2007)
Shewanella oneidensis MR-1 H-NOX regulation of a histidine kinase by nitric oxide. Biochemistry 46: 13677–13683.
Rao, F., Qi, Y., Chong, H.S., Kotaka, M., Li, B., Li, J., et al.
(2009) The functional role of a conserved loop in EAL
domain-based cyclic di-GMP-specific phosphodiesterase.
J Bacteriol 191: 4722–4731.
Ren, T., Zamboni, D.S., Roy, C.R., Dietrich, W.F., and Vance,
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942
942 H. K. Carlson, R. E. Vance and M. A. Marletta !
R.E. (2006) Flagellin-deficient Legionella mutants evade
caspase-1- and Naip5-mediated macrophage immunity.
PLoS Pathog 2: e18.
Roberts, G.P., Youn, H., and Kerby, R.L. (2004) CO-sensing
mechanisms. Microbiol Mol Biol Rev 68: 453–473.
Rodionov, D.A., Dubchak, I.L., Arkin, A.P., Alm, E.J., and
Gelfand, M.S. (2005) Dissimilatory metabolism of nitrogen
oxides in bacteria: comparative reconstruction of transcriptional networks. PLoS Comput Biol 1: e55.
Shatalin, K., Gusarov, I., Avetissova, E., Shatalina, Y.,
McQuade, L.E., Lippard, S.J., and Nudler, E. (2008) Bacillus anthracis-derived nitric oxide is essential for pathogen
virulence and survival in macrophages. Proc Natl Acad Sci
USA 105: 1009–1013.
Stone, J.R., and Marletta, M.A. (1996) Spectral and kinetic
studies on the activation of soluble guanylate cyclase by
nitric oxide. Biochemistry 35: 1093–1099.
Sudhamsu, J., and Crane, B.R. (2009) Bacterial nitric oxide
synthases: what are they good for? Trends Microbiol 17:
212–218.
Tamayo, R., Tischler, A.D., and Camilli, A. (2005) The EAL
domain protein VieA is a cyclic diguanylate
phosphodiesterase. J Biol Chem 280: 33324–33330.
Tamayo, R., Pratt, J.T., and Camilli, A. (2007) Roles of cyclic
diguanylate in the regulation of bacterial pathogenesis.
Annu Rev Microbiol 61: 131–148.
Thormann, K.M., Saville, R.M., Shukla, S., and Spormann,
A.M. (2005) Induction of rapid detachment in Shewanella
oneidensis MR-1 biofilms. J Bacteriol 187: 1014–1021.
Thormann, K.M., Duttler, S., Saville, R.M., Hyodo, M.,
Shukla, S., Hayakawa, Y., and Spormann, A.M. (2006)
Control of formation and cellular detachment from
Shewanella oneidensis MR-1 biofilms by cyclic di-GMP.
J Bacteriol 188: 2681–2691.
Tischler, A.D., and Camilli, A. (2004) Cyclic diguanylate (c-di-
GMP) regulates Vibrio cholerae biofilm formation. Mol
Microbiol 53: 857–869.
Tuckerman, J.R., Gonzalez, G., Sousa, E.H., Wan, X., Saito,
J.A., Alam, M., and Gilles-Gonzalez, M.A. (2009) An
oxygen-sensing diguanylate cyclase and phosphodiesterase couple for c-di-GMP control. Biochemistry 48:
9764–9774.
Wan, X., Tuckerman, J.R., Saito, J.A., Freitas, T.A., Newhouse, J.S., Denery, J.R., et al. (2009) Globins synthesize
the second messenger bis-(3′-5′)-cyclic diguanosine
monophosphate in bacteria. J Mol Biol 388: 262–270.
Wang, Y., Dufour, Y.S., Carlson, H.K., Donohue, T.J., Marletta, M.A., and Ruby, E.G. (2010) H-NOX-mediated nitric
oxide sensing modulates symbiotic colonization by Vibrio
fischeri. Proc Natl Acad Sci USA 107: 8375–8380.
Yamamoto, Y., Klein, T.W., and Friedman, H. (1996)
Immunoregulatory role of nitric oxide in Legionella
pneumophila-infected macrophages. Cell Immunol 171:
231–239.
Zafiriou, O.C., McFarland, M., and Bromund, R.H. (1980)
Nitric oxide in seawater. Science 207: 637–639.
Zhao, Y., Brandish, P.E., Ballou, D.P., and Marletta, M.A.
(1999) A molecular basis for nitric oxide sensing by soluble
guanylate cyclase. Proc Natl Acad Sci USA 96: 14753–
14758.
Supporting information
Additional supporting information may be found in the online
version of this article.
Please note: Wiley-Blackwell are not responsible for the
content or functionality of any supporting materials supplied
by the authors. Any queries (other than missing material)
should be directed to the corresponding author for the article.
© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 77, 930–942